When AI gains awareness:

AI is advancing rapidly, for example, GPT 4 can pass the bar exam and complex medical exams. Kim Kardashian had to take the bar exam three times before passing on the fourth try. However, we wonder what happens when AI becomes conscious and whether it already is without our knowledge. We already treat language models with empathy because they seem human-like. We find ourselves being considerate and polite towards them, just like we would with NPCs in video games. This raises concerns about ethics, safety, and society. Before catastrophe strikes, we need to understand consciousness. It is the state of being awake, aware, and everything making sense. When under anesthesia or deep in a dream, we are not conscious. People might think that large language models like chat GPT are conscious because they are intelligent, but that's not true.

We need to understand what exactly clicks in our brains and replicate it. It will take a long time and a lot of energy, but it's theoretically possible. However, if the soul perspective is correct, making AI conscious could be very difficult or even impossible. It's like saying machines might need a dash of soul to become aware. Additionally, determining consciousness is subjective, as only you can say if you're conscious or not. If AI becomes so good at pretending to understand and think like us, it would be hard for us to distinguish if it's conscious or not. For example, Google's AI chatbot Lambda stirred up a debate when it claimed to think and feel. Google maintains that Lambda isn't sentient and is just mimicking humans. The situation has prompted scientists to develop a checklist to test self-awareness in AI, but so far, none of the language models have met the criteria. Testing consciousness in other beings, such as animals and newborns, is also being considered. However, these studies will take many years, and it would be awkward if AI accidentally becomes aware while we're still trying to understand it.

Robots might start to ask themselves what their purpose is, if there is a possibility that sentient AI can have empathy and emotional intelligence. Remember Frankenstein's monster, that it came to life wasn't the problem but that it could feel things and if we can find a way to coexist in peace now that we'd have a moral responsibility to take care of these machines we'd have to figure out what rights to give them and if we had the right to turn them off and it would be very difficult to coexist in peace with these AIs considering the fact that humans can't even be at peace with each other. They can try their luck at peace with us though and let's hope the AI ​​decides to be better because if it decides it doesn't like us so much it could be so violent therefore, we'd have to make sure those interests don't clash with ours.

All Skynet style very dangerous situations and singularity which could be a hypothetical time in the future, when the technology is cool enough to start improving we will start our technological development again. All these things can cause some real danger to our brains and we are just not ready for this kind of situation right now which is why we have to be very careful but all action must be taken, we must invest in research to better understand the mind of not only AI but also medical law and Animal Welfare. We also need to dig into the social sciences and humity to figure out what to do when AI seems to be self-aware. The association of Mathematical Cognitive Sciences sent an open letter to AI creators saying that although we shouldn't stop with the AI research still we need to slow down if we really want to better understand insights about the AI self awareness especially when it comes to AI because everything is changing so fast and there might not be time to react if anything were to happen.

Anyway, the world of artificial intelligence is getting really interesting with big questions about self-awareness, intelligence, and what makes us human. But no need to worry too much, creating AI consciousness is still a long way off. Despite some saying we're close due to recent progress, right now, we only have basic AI and some language models. Most experts in AI think we won't have machines that are like humans until at least the 2050s. So, just relax and keep an eye on the news for updates.